Geo-visualization of Wikipedia Environmental Issues

This interactive map allows you to visualize environmental issues available in Wikipedia by topic and date.

Briefly summarized, these issues and their affiliate topics were obtained by querying the Wikipedia knowledge base. More specifically, we used the Wikipedia knowledge base dump up to 2016. The concepts were obtained by applying automatic clustering techniques to group together issues based on their graph similarity. These clusters were then manually reviewed, filtered, cleaned and labelled with a topic. Locations were obtained by linking phrases in the short abstracts to concepts in the Wikipedia, filtering for locations, then querying the Geonames knowledge base for geo-coordinates. Finally, dates were obtained also from the short abstracts using Named Entity Recognition.

We obtained over 15000 geo-located issues. About 12000 of these are Biota (e.g. endangered plants or animals). The rest are issues ranging from air pollution, climate change, oil spills and other disasters.